On March 28, Yannis Pantazis, PhD, gave an invited talk at the Scaling Cascade in Complex Systems workshop held at the Free University of Berlin. The title of the talk is “Information-theoretic Uncertainty and Sensitivity Bounds for Stochastic Dynamics and Rare Events“. More info at:
The preprint version of our new publication is now available on bioRxiv. This work applies two distinct causal discoveries methods on mass-cytometry (CyTOF) data, producing causal findings that are reproducible across two independent studies. Check the full text here: https://t.co/JmW59B3nmt
Following our annual tradition we celebrated the coming of the new year by “cutting” the cake, better known as Vasilopita.(https://en.wikipedia.org/wiki/Vasilopita).
This year, Klio was lucky one !
Happy New Year MxMers!
Prof. Tsamardinos visited Athens during 24-25 of November and gave two invited talks about his scientific work. The first one was held on the 24th of November, about: “Advances in Feature Selection in Data Analytics”, at the Department of Digital Systems of University of Piraeus and the second one on the 25th of November about:” Logic-Based Causal Discovery for Heterogeneous Datasets”, which was held in the Department of Informatics and Telecommunications, National and Kapodistrian University of Athens.
We are looking for an enthusiastic, highly motivated software developer with a strong background in full-stack web applications to join our team and be the lead programmer for our data analysis web platform. The platform will host applications on the cloud providing data analytics solutions for researchers and practitioners across disciplines.
As a software developer, you will design and implement a performant, scalable, user friendly, high availability web-based environment, and contribute to the code for all parts of its life cycle from design, development, and deployment, to maintenance, operation, and support.
The software development in which the successful applicant will be involved has been the result of many years of our team’s leading research in the fields of machine learning and artificial intelligence for data analysis. At (mensxmachina) we go beyond the data and develop cutting-edge algorithms and software for big-data predictive analytics. We have pioneered technology for scalable, accurate, and principled variable selection methods and probabilistic network learning, attracting several national and EU grants.
Skills and Qualifications
- Bachelor’s degree in computer science or related field
- At least 2 years’ experience in developing full-stack web applications using the following technologies
- Knowledge of software engineering practices (e.g. design patterns, unit testing), and of version control tools (most preferably Git) and build automation tools (most preferably Maven)
- Excellent communication skills
- Professional working proficiency in English
- Ability to work within an interdisciplinary team
- Independent thinking, ability to work autonomously, coming up with proposals and solutions, taking initiatives
- Master’s degree in computer science or related field
- Experience in the following technologies
- REST/SOAP, NoSQL databases, Visualization tools (e.g. Kibana, Google charts), Bootstrap, JPA Hibernate
- Cloud services technologies like Amazon Web Services
- High-performance computing services
- Scripting languages such as Python and Bash
- Scientific computing languages such as R and MATLAB
- Experience in communicating and working with experts (e.g. in a science community) in a co-development model
The compensation is highly competitive with respect to similar positions. The salary will be tailored to the experience and capacity of the candidate.
Applications will be continuously evaluated every two weeks and until the position is filled. The first round will end on November 11th, 2016. Interested candidates can send an email to mensxmachina at gmail dot com with the following information:
- A cover letter stating their motivations for applying
- A detailed CV
Prof. Ioannis Tsamardinos gave a Distinguished Lecture on August 19th about: “Logic-Based Causal Discovery for Heterogeneous Datasets,” as part of the Distinguish Lecture Series organized by “Center for Causal Discovery”, University of Pittsburgh, Carnegie Mellon University, Pittsburgh Supercomputing Center and Yale University.
On the 26th of August 2016 Prof. Ioannis Tsamardinos introduces the novel field of Logic-based Integrative Causal Discovery in an invited lecture at the North Carolina State University, host Prof. Yannis Viniotis
Link to the talk: video
Our post doc, Sofia Triantafillou, gave a tutorial on integrative, logic-based causal discovery in the prestigious conference “Uncertainty in Artificial Intelligence” (UAI 2016, July 25-29, NY). Take a look at the slides here (link), or wait until the video is available on UAI’ s webpage: http://www.auai.org/uai2016/tutorials.php
Prof. Tsamardinos talk at LINCS (Laboratory of Information, Networking and Communication Sciences), Paris, France 24/02/2016.
Abstract: Computational Causal Discovery aims to induce causal models,causal networks, and causal relations from observational data withoutperforming or by performing only few interventions (perbutations,manipulations) of a system. While predictive analytics create models thatpredict customer behavior for example, causal analytics create models thatdictate how to affect customer behavior. A recent approach to causaldiscovery, which we call logic-based integrative causal discovery, will bepresented. This approach is more robust to statistical errors, makes morerealistic and less restrictive assumptions (e.g., admits latent confoundingfactors and selection bias in the data) and accepts and reasons withmultiple heterogeneous datasets that are obtained under different samplingcriteria, different experimental conditions (perbubations, interventions),and measuring different quantities (variables). The approach significantlyextends causal discovery based on Bayesian Networks, the simplest causalmodel available, and is much more suitable for real business or scientificdata analysis.